USF Libraries
USF Digital Collections

Racial equity in exclusionary discipline practices

MISSING IMAGE

Material Information

Title:
Racial equity in exclusionary discipline practices
Physical Description:
Book
Language:
English
Creator:
Tremper, Mary M
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
race
suspension
socioeconomic status
behavior management
adolescents
Dissertations, Academic -- School Psychology -- Specialist -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: The present study examined whether external (out-of-school) suspensions are applied equitably to students of different ethnic backgrounds who commit violent and nonviolent offenses. The hypotheses presented in this study were addressed through secondary analysis of disciplinary records from a large metropolitan school district in Florida. The results indicate that, for the group of 1,667 tenth grade students included in this analysis, racial equity was related to the type of offense, as well as to the student's socioeconomic status. Racial differences were found when SES was not considered, with African American students more likely to be suspended from school for status offenses and violent offenses. The same degree of racial disproportionality was not found among low SES students.However, middle and higher SES students appeared to account for much of the racial disproportionality seen in the sample, with African American students in this group more likely to be suspended for both violent and status offenses.
Thesis:
Thesis (Ed.S.)--University of South Florida, 2004.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Mary M. Tremper.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 73 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001498132
oclc - 57717271
notis - AJU6727
usfldc doi - E14-SFE0000535
usfldc handle - e14.535
System ID:
SFS0025226:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Racial Equity in Exclusionary Discipline Practices by Mary M. Tremper A thesis submitted in partial fulfillment of the requirement s for the degree of Education Specialist Department of Psychological and Social Foundations College of Education University of South Florida Major Professor: Kelly Powell-Smith, Ph.D. Members: Linda Raff aele Mendez, Ph.D. Robert Dedrick, Ph.D. Date of Approval: November 5, 2004 Keywords: suspension, race, adolescents, behavior management, socioeconomic status Copyright 2004, Mary M. Tremper

PAGE 2

i Table of Contents List of Tables iii Abstract v Introduction 1 Rationale 2 Research Questions and Hypotheses 3 Review of Literature 8 The Nature of Disproportionality 8 Demographic Trends in Suspension 9 Sources of Disproportionality 10 Reasons for Suspension 13 Problems and Attitudes Associated with Suspension 15 Summary 17 Method 20 Participants 20 Measures 25 Procedures 29 Obtaining the data 29 Data integrity 29 Type of analysis 30 Assumptions of Chi Square 31 Results 32 Research Question One 35 Research Question Two 37 Research Question Three 39 Research Question Four 41 Analysis at the School Level 43 Discussion 51 Research Question One 51 Research Question Two 52 Research Question Three 53 Research Question Four 54

PAGE 3

ii Comparison Within Schools 55 Summary and Implications 56 Limitations of this Study 58 Directions for Future Research 59 Implications for School Psychology Practitioners 61 References 64

PAGE 4

iii List of Tables Table 1 Numbers of Caucasia n and African American Students in the 2000-2001 Tenth Grade Sample with O ne Referral 21 Table 2 Free/Reduced Lunch Status of Students in the 2000-2001 Tenth Grade Sample with One Referral 22 Table 3 Frequency of Students Per Number of Referrals 23 Table 4 Referral Codes and Categories 28 Table 5 Referral Code Frequencies fo r all Tenth Graders 33 Table 6 Referral Code Categories for Students with One Referral 35 Table 7 Frequencies of Suspensions for Five Categories of Referrals, by Socioeconomic Status 36 Table 8 Chi-Square Analysis for Res earch Question One 37 Table 9 Frequencies of Suspensions for Five Categories of Referrals, by Race 38 Table 10 Chi-Square Analysis for Re search Question Two 39 Table 11 Frequencies of Suspensions of Low SES Students for Five Categories of Refe rrals, by Race 40 Table 12 Chi-Square Analysis for Res earch Question Three 41 Table 13 Frequencies of Sus pensions of Middle and Higher SES Students for Five Categories of Referrals, by Race 42 Table 14 Chi-Square Analysis for Res earch Question Four 43 Table 15 Frequency and Percentage of Single Referrals, by Site 44

PAGE 5

iv Table 16 Frequency and Percentage of Suspensions for Referral Categories by Race, High School A 45 Table 17 Frequency and Percentage of Suspensions for 5 Referral Categories by Race, High School B 46 Table 18 Frequency and Percentage of Suspensions for 5 Referral Categories by Race, High School C 47 Table 19 Frequency and Percent age of Suspensions for 5 Referral Categories by Race, High School D 48 Table 20 Percentages of C aucasian and African American Students Suspended for Each Referral Code 49

PAGE 6

v Racial Equity in Exclusionary Discipline Practices Mary M. Tremper ABSTRACT The present study examined whether external (out-ofschool) suspensions are applied equitably to students of di fferent ethnic backgrounds who commit violent and nonviolent offenses. The hypot heses presented in this study were addressed through secondary analysis of disciplinary records from a large metropolitan school district in Florida. The results indicate that, for t he group of 1,667 t enth grade students included in this analysis, racial equity was related to the type of offense, as well as to the student’s socioeconomic status Racial differences were found when SES was not considered, with African American students more likely to be suspended from school for status off enses and violent offenses. The same degree of racial disproportionality was not found among low SES students. However, middle and higher SES students appeared to account for much of the racial disproportionality seen in the sa mple, with African American students in this group more likely to be suspended fo r both violent and status offenses.

PAGE 7

1 Chapter I Introduction In today’s public schools, disciplinary measures that involve exclusion of students from the educational setting (e .g., suspension and expulsion) are increasingly common (Johnst on, 2000). The current em phasis on zero-tolerance disciplinary practices has led to renewed debate over the fairness and effectiveness of such procedures. A typical argument in favor of suspens ion is that, “The majority of students who are interested in learning should not suffer the constant disruption of the very few,” (Wu, Pink, Crain, & Moles, 1982). This sentiment has been echoed by administrators, who tend to view suspens ion and expulsions not as interventions designed to help the affected student, but as measures to preserve an orderly educational environment (Bowditch, 1993). Federal law mandates that all st udents have access to a “free and appropriate education” in the least restri ctive environment possible (IDEA, 1997). Because external discipline involves the removal of affected students from educational resources and social networks, equity issues related to its application have caused controversy among educators and researchers. Examination of the demographics of suspended students reveals that African American students, particularly males, are overrepresented in this group. European American males come in second, followed by African American females, while European

PAGE 8

2 American females are suspended at the lo west rate (Office for Civil Rights, 1994). Rationale Exclusionary discipline has been a simultaneously commonplace and controversial educational practice for many years. A number of empirical studies have described racial and gender disparit ies in the application of such consequences. Suspension from school is synonymous with denial of educational resources. In addition, suspension (particu larly multiple external suspensions) is associated with negative trajectories leading to dropout, delinquency, and poor educational achiev ement (Bowditch, 1993; Rodney, Crafter, Rodney, & Mupier, 1999; Sprague, Walker, Stieber, Simonsen, Nishioka, & Wagner, 2001). Therefore, it is imperative that educ ators ensure that, when such consequences are used, t hey are equitably applied. The extant literature has established that Afric an American students are suspended from school at higher rates than their European American peers. However, it is not known whether this disparity is due to student behavior, differential referral practices among teac hers, or administrative bias in the application of disciplinary procedures. The present study examined whether internal (e.g., in-school) and external (e.g., out-of-school) suspensions are applied equitably to students of different ethnic backgrounds who commit violent and nonviolent offenses. Although students receive disciplinary re ferrals for many different specific types of offenses, the infractions can be classified into five more general

PAGE 9

3 categories (Bowditch, 1993). Using this cl assification system allows a lengthy list of district-wide referral codes to be si mplified and more readily conceptualized. Property offenses include such propertybased acts as steali ng and vandalism. Status offenses include truancy and disres pect. These types of offenses are nonviolent and involve violati ons of social mores or rules specific to the school setting rather than illegal acts. The controlled substances category encompasses possession of alcoho l, tobacco, and other unauthorized substances on school property. Violent o ffenses include fighting, possession of weapons, and assault. The “other” category encompasses referrals for which the exact offense code was not recorded. The hypotheses presented in th is study were addressed through secondary analysis of disciplinary records from one school district in southwest Florida. By nature of being a seco ndary analysis, this study has certain limitations. The author did not have cont rol over the creation or coding of the data set. However, secondary analysis al lows for efficient and comprehensive study of a relatively large sample. Research Questions and Hypotheses Based on the current literature rela ted to racial equity in exclusionary discipline practices, several research ques tions were formulated to address the concerns of this study. Research Question 1. Are students identified as having low socioeconomic status (SES) based on free/reduced lunc h status suspended from school more

PAGE 10

4 frequently than students not identified as having low SES with disciplinary referrals for the same types of offenses? Researcb Question 2. Are African American st udents suspended from school more frequently than Caucasi an students with disci plinary referrals for the same types of offenses? Research Question 3. Are low SES African Amer ican students suspended from school more frequently than low SES Ca ucasian students with disciplinary referrals for the same types of offenses? Research Question 4. Are middle and higher SES African American students will be suspended from school more fr equently than middle and higher SES Caucasian students with discipl inary referrals for the sa me types of offenses? The following hypotheses were generated based on current research findings related to exclusionary discipline: Hypothesis 1. Students identified as having low socioeconomic status (SES) based on free/reduced lunch status will be suspended from school more frequently than students not ident ified as having low SES. a. Low SES students will be suspended from school more frequently than middle and higher SES students for violent offenses. b. Low SES students will be suspended from school more frequently than middle and higher SES students for property offenses.

PAGE 11

5 c. Low SES students will be suspended from school more often than middle and higher SES student s for status offenses. d. Low SES students will be suspended from school more frequently than middle and higher SES students for offenses involving controlled substances. e. Low SES students will be suspended from school more frequently than middle and higher SES students for offenses categorized as “other.” Hypothesis 2. African American students will be suspended from school more frequently than Caucasian students. a. African American students will be suspended from school more frequently than Caucasian students for violent offenses. b. African American students will be suspended from school more frequently than Caucasian students for property offenses. c. African American students will be suspended from school more frequently than Caucasian studen ts for status offenses. d. African American students will be suspended from school more frequently than Caucasian students for offenses involving controlled substances. e. African American students will be suspended from school more frequently than Caucasian students for offenses categorized as “other.” Hypothesis 3. Low SES African American students will be suspended from school more frequently than low SES Caucasian students.

PAGE 12

6 a. Low SES African American studen ts will be suspended from school more frequently than low SES Caucasian students for violent offenses. b. Low SES African American st udents will be suspended from school more frequently than low SES Caucasian students for property offenses. c. Low SES African American studen ts will be suspended from school more frequently than low SES Caucasian students for status offenses. d. Low SES African American st udents will be suspended from school more frequently than low SES Caucasian students for offenses involving controlled substances. e. Low SES African American studen ts will be suspended from school more frequently than low SES Caucasian students for offenses categorized as “other.” Hypothesis 4. Middle and Higher SES Afri can American students will be suspended from school more frequently than middle and higher SES Caucasian students. a. Middle and higher SES African American students will be suspended from school more frequently than middle and higher SES Caucasian students for violent offenses. b. Middle and higher SES African American students will be suspended from school more frequently t han middle and higher Caucasian students for property offenses.

PAGE 13

7 c. Middle and higher SES African American students will be suspended from school more frequently than middle and higher SES Caucasian students for status offenses. d. Middle and higher SES African American students will be suspended from school more frequently than mi ddle and higher SES Caucasian students for offenses involv ing controlled substances. e. Middle and higher SES African American students will be suspended from school more frequently than middle and higher SES Caucasian students for offenses cat egorized as “other.”

PAGE 14

8 Chapter II Review of Literature The chapter reviews recent research on the issue of racial equity in exclusionary discipline practices in secondary schools. First, demographic trends in the use of suspension and reas ons for suspension are discussed. Next, student attitudes and problem s associated with suspension are summarized. Finally, limitations of the extant research a nd the need for further investigation are addressed. The Nature of Disproportionality Racial disproportionality has long been a “hot topic” in such areas as special education and school discipline. Rec ent research on the interpretation of disproportionality data points out that the two most co mmon methods of reporting such data can yield very different percentages (MacMillan & Reschly, 1998; Reschly, 1997). One common method com pares the total proportion of the target group (e.g., percentage African Am erican students) in the population as a whole with the proportion of t he target group in the cat egory of interest (e.g., percentage of African American student s suspended from school). The other method simply examines the percentage of a population (e.g., African American students) in the category of interest (e.g., students suspended from school). Disproportionality studies hi storically have used inconsistent criteria in determining whether a statistical discrepancy represents a significant overrepresentation. In one common st andard, a group is considered to be

PAGE 15

9 overrepresented in a target category if it s representation in the target category exceeds its representation in the population by 10% or more (Reschly, 1997). Demographic Trends in Suspension Exclusionary discipline, such as su spension, is an increasingly popular alternative for administrator s at the middle and high sch ool levels. Federal data indicate that 6.9% of public school st udents in the United States were suspended from school at least once in 1998. This figure represents an increase from 3.7% in 1974 (Johnson, 2000). Suspension is employed with greater frequency as students progress from elementary to middle to high school (Costenbader & Marksson, 1994). Across studies of suspension in middle and high schools, one consistent trend is that African American students, particularly males, are suspended in numbers significantly dispr oportionate to their total enr ollment (e.g., Wu, Pink, Crain, & Moles, 1982). Racial disparit ies in the use of suspension have been extensively documented at local and nationa l levels. A survey of 43,034 public schools in 4,692 school districts, encompas sing more than 25 million students (Office for Civil Rights, 1994) revealed 1,524,241 cases of suspension during one school year. An examination of like lihood ratios derived from that survey indicated that African Am erican boys were twice as likely to be externally suspended as Caucasian boys, and six times as likely to be suspended as Caucasian girls (Gregory, 1997). In a survey examining suspensio n rates in middle and high schools representing rural, suburban, and ur ban areas in 10 states, Costenbader and

PAGE 16

10 Markson (1994) found that African Amer ican students were suspended in greater proportion than other students, while Asian American students were underrepresented in suspensions. In an investigation of school disciplin e practices in middle schools in two Midwestern cities, Skiba, Peterson, and Williams (1997) found similarly striking ethnic disproportionalities, with Native American students receiving the most disciplinary referrals and suspensions, fo llowed by African American students. Other factors associated with higher referral rates included being male and qualifying for free or reduced lunch. In addition, students labeled Emotionally Handicapped received more office refe rrals and suspensions than students in general education and student s in other special education programs. Students identified as having learning disabilities or mild ment al handicaps also were suspended more frequently than students not receiving any special education services. Sources of Disproportionality Researchers have posited various r easons for the widespread racial disproportionality in exclus ionary discipline. Thes e causes have ranged from statistical artifacts to t he depiction of schools as prim arily white, middle-class matriarchies in which African American ma les of lower socioeconomic status are at a cultural disadvantage (Gregory, 1997). Research t hat empirically examines the reasons for racial disparities in sus pension is the most useful for determining possible ways to remediate this problem.

PAGE 17

11 In a seminal analysis of national suspension data, Wu et al. (1983) investigated the relation between ty pes of student misbehavior, teacher judgments and attitudes, admin istrative structures, t eacher perceptions of academic potential, racial bias and sus pension rates. The analysis revealed a complex interplay among factors related to the operation of the school, ultimately characterizing suspension more as a refl ection of school culture than of student behavior. Suspension was more prevalent in schools exhibiting the following characteristics: (1) students view the sc hool’s governance as largely unfair or inconsistent, (2) students view the teachers as uninterested in them, (3) teachers view students as having poor problem-solvi ng skills, (4) disciplinary matters are primarily handled through administrativ e channels, and (5) racial and academic biases are prevalent in the school. Fu rther, the overrepres entation of nonwhite students among those suspended from school was relatively independent of variables commonly posited as mitigati ng factors. When differences in socioeconomic status were statistically controlled, nonwhite students still were suspended in numbers disproportionate to their Caucasian peers across most settings. In addition, suspension rates di d not differ significantly for African American students attending schools with more Caucasian or more nonwhite teachers. In a recent position paper incorporat ing disciplinary incident data from a large, Midwestern public school district Skiba, Michael, Nardo, and Patterson (2000) investigated possible sources of racial disproportionality in suspension rates. They examined w hether the overrepresentati on of African American

PAGE 18

12 students could be explained as an artifact of statistical methodology, a reflection of socioeconomic differences, or a produc t of higher rates of misbehavior among African American students. Based on the presence of significant racial discrepancies in the number of suspensions, regardless of whether st atistical methodology focused on the proportion of suspended students or the “ten percent of t he population standard,” Skiba et al. (2000) concluded that racial disproportionality in suspension does not represent an artifact of statistical methodol ogy. Echoing the findings of Wu et al. almost two decades earlier, Skiba et al (2000) concluded that, while students of low socioeconomic status are suspended from school more frequently than highSES peers, disciplinary outcomes and race remain linked when SES was statistically controlled. Finally, Skiba et al. (2000) addressed the issues of student behavior and referral bias. Citing research on disp roportionate represent ation of minority students in special education (Serwatk a, Deering, & Grant, 1995) and the overrepresentation of Afric an Americans in the criminal justice system (e.g., New York State Attorney General’s Office Civil Rights Bureau, 1999), the authors concluded that high suspension rates among African American students are most likely the result of referr al bias, in which African American students are more likely than Caucasian students to be referr ed to administrator s for disciplinary action. The authors posited that African Americans ar e overrepresented in the criminal justice system because this popula tion is more likely to be contacted by

PAGE 19

13 law enforcement (e.g., “racial profiling”), and that a similar effect occurs in school discipline. The investigation conducted by Ski ba et al. (2000) provides a valuable analysis of exclusionary discipline prac tices in a large metropolitan school district, and explores some common hypothese s related to this issue. However, the issue of referral bias was not empi rically addressed. Also, the paper does not address whether Caucasian and Afri can American students were suspended for the same types of offenses. The overrepresentation of Afri can American students among those suspended and expelled from school has been increasingly problematic as “zero tolerance” disciplinary policies have c aused suspension rates to skyrocket. The impact of zero tolerance on minority students has been addressed in recent testimonies before the U. S. Commission on Civil Right s. As the use of exclusionary discipline increases, so does the overrepresentation of African American students in suspension and ex pulsion (Advancement Project, 2000; Keleher, 2000). Keleher has called for sc hool districts and states to collect comprehensive discipline data and set m easurable goals toward reducing the use of exclusionary discipl ine and eliminating raci al disproportionalities. Reasons for Suspension In the analysis conducted by Costenbader and Markson (1994), the most common reason for suspension was physica l aggression, accounting for 35% of internal suspensions and 51% of exte rnal suspensions in middle schools and 12% of internal suspensions and 33% of external suspensions in high schools.

PAGE 20

14 The findings from the middle school study by Skiba et al. (1999) were similar to those from Costenbader and Markson’s survey in that the most common disciplinary infraction leading to suspens ion was fighting. Office referral data were analyzed in the middle school study, and indicated that the most common behaviors leading to referrals were noncompliance and disrespect. For infractions other than fighting, no cons istent relationship was found between the type of offense and the adminis trative consequences. Similar findings regarding reasons for suspension have been echoed in other school districts. McFadden, Marsh, Price, and Hwang (1992) examined discipline records for 4,391 students receiv ing disciplinary action in a Florida school district between August 1987 and Ap ril 1988. The researchers grouped the 25 disciplinary offense cat egories identified by the district into five levels of severity, with the most serious infracti ons involving drugs, weapons, or violence, and the least serious involving such behavio rs as “bothering others” (p. 143). Again, African American students received the most office referrals and suspensions in proportion to their tota l enrollment, accounting for 22% of all students in the district, 36.7% of disci plinary referrals, 43 .9% of external suspensions, and 23% of internal suspens ions. However, this study did not examine possible links between race and the types of offenses for which students were suspended from school. In summary, the ex tant research on racial equity in exclusionary discipline indicates that African American student s are overrepresented among suspended students. The studies discussed in this chapter provide thorough documentation

PAGE 21

15 that racial disproportionality in the application of exclus ionary discipline is a real and long-standing problem. However, the reasons for this overrepresentation remain unclear. The extant research does not address the issue of whether students from ethnic minority groups rece ive referrals and suspensions for the same types of misbehavior, or whether exclusionary discipli ne is differentially applied to various ethnic groups. Problems and Attitudes Associated with Suspension Suspension from school is a risk fact or for negative outcomes, including grade retention and dropout. In a multiple regression study examining predictors of grade retention among Af rican American adolescent males, Rodney, Crafter, Rodney, and Mupier (1999) f ound that the number of suspensions from school was the strongest predictor of grade retent ion. The other identified predictors (conduct disorder and home disciplinary prac tices) are factors largely outside the control of educators. Suspension appears to place students already having problems at additional risk for academic failure by excluding them from the educational environment. Costenbader and Markson (1998) su rveyed urban and rural secondary students about their experiences and attit udes related to suspension and found that students who had been suspended were more likely than others to be involved with the legal system. Regardi ng their feelings about suspension, the majority of students report ed feeling either angry at the person who suspended them or happy to get out of the school situation. The authors theorize that suspension places students at increased risk for delinquency because it removes

PAGE 22

16 them from positive social networks and increases thei r potential contact with the “delinquent subculture” (p. 73). Perhaps the most insidious threat suspension poses to adolescents relates to alienation from the school environment and adoption of a “troublemaker” identity (Bowditch, 1993). In an ethnographic study of disciplinary procedures at an inner city high school Bowditch (1993) found that, once a student was referred for administrative disci pline, whether or not to suspend was generally based on the student’s prior disci plinary history (i.e., previous suspension predisposes a student to later suspension), and the degree of remorse shown by the student. Thus, the same students were often repeatedly excluded from the educational setting, and the suspension was not applied in a consistent manner to all students. Bo wditch argues that based on the racial disproportionality evident in school disciplinary practices, suspension serves to perpetuate racial and class stratifica tion in the larger society. In another investigation of middle and high school suspensions, MorganD’Atrio, Northup, LaFleur, and Spera ( 1996) found that students with recurrent suspensions tend to exhibit considerable academic and social skills deficits. Morgan D’Atrio et al. used the school-wide data processing program to examine suspension rates and identify students with multiple suspensions. In this descriptive study, individual assessment was conducted to identify potential deficits in social or academic skills. Measures used include a semi-structured interview, standardized readi ng assessments, the Social Skills Rating Systemstudent and teacher reports (Gresham & Elliot, 1990), and the Child Behavior

PAGE 23

17 Checklistself-report and teacher measures (Achenbach & Edelbrock, 1991). More than half of the student s with multiple suspensions scored more than two grade levels below placement on a word recognition task, and 33% of the middle school students were rated by their teac hers as having significant social skills deficits. Students with repeated suspens ions had a mean grade point average (GPA) below 2.0. Students typically perceive suspension as punishment (Miller, 1986). Literature in the field of applied behavior analysis has documented numerous undesirable consequences of punishment. For example, when an administrator applies a punishment-based procedure su ch as suspension and the student ceases to engage in the negative behavior, t he administrator is reinforced by the termination of that behavior (Cooper, Her on, & Heward, 1987). He or she may be more likely to suspend that student or other students for futu re misbehavior. In addition, punishment may lead to an emot ional or aggressive response on the part of the student. Avoidance or e scape behaviors also may emerge, with students becoming truant or viewing the actual period of suspension as a welcome escape from aversive in teractions with school personnel. Summary Research on exclusionary disci pline indicates clear racial disproportionality, although t he reasons behind this issue remain unclear due to limitations of the extant re search on this topic. The research to date, however, indicates that African Amer ican students, especially boys, are more likely to be suspended from school than their peers. The disproportionality could be due to

PAGE 24

18 actual behavioral differences among studen ts, referral bias at the classroom level, disciplinary bias at t he administrative level, or some combination of factors. The preponderance of empirical studies on racial equity in suspension have clearly documented the existence of wi despread, long-standing disparities but not the reasons for them. The disproportion ality is not likely to be fully accounted for by statistical methodology, socioecon omic differences, or documented group differences in behavior. While referra l bias has been documented on a limited basis (Huberty, 1994), the relationshi p between race and the disciplinary outcomes of referrals has not been directly examined. This issue has not been addressed empirically since the advent of ze ro tolerance disciplinary practices in the late 1990’s. It can be argued that suspension from school places already at-risk students at even greater risk for academic failure and negative social outcomes by limiting their connection to teachers, positive peer influences, and educational opportunities. Students who experience excl usionary discipline are more likely to suffer academic and social skills defic its, grade retention, dropout, and delinquency. Suspension appears to a widel y-employed practice that conflicts with the notion that interventions are va luable only when associated with positive outcomes for students (Reschly & Tilly, 1993). Considering the long and stable histor y of the overrepresentation of African American students among those sus pended from school, it is important for practitioners to address this issue fr om a problem-solving perspective, rather than merely documenting continued inequa lity. By adopting a problem-solving

PAGE 25

19 approach, educators can advocate more effe ctively for students who continue to be impacted by exclusionary discipline. In vestigation of the types of offenses that most often lead to the suspens ion of African American students can be linked to the development of interventions that target those specific behaviors on the classroom, school, and system-wide leve ls. The present study will provide information about the impact of equity issues when consequences are doled out at the administrative level.

PAGE 26

20 Chapter III Method Participants Participants were students involved in a longitudinal study administered by a large metropolitan Florida public school system. The longitudinal study has collected data annually from the ent ire cohort of students who entered kindergarten in the fall of 1989. The initial cohort was comprised of 8,268 students. Demographi c features of the sample us ed in the present study are summarized in Table 1. Parents, teachers, and/or st udents have been surveyed yearly using scales and questionnaires co vering a wide range of issues in the family, behavioral, and academic domains. T he specific variables of interest and the sample of participating students surv eyed varied each year. A committee of researchers and school personnel convened yearly to identify areas of concern and select questions for the survey. The committee membership and areas of interest have changed from year to year. The longitudinal study’s database was updated yearly with information from the general stude nt data file. This informa tion included grades, special education status, discipline referrals, and standardized achievement test scores. For the present study, disciplinary reco rds from the general student data files for the 1999-2000 school year were analyzed. The cohort participating in the longitudinal study was selected for this in vestigation because of the large sample size, and the availability of data beyond the general student data, which could be

PAGE 27

21 included in later analyses related to t he same problem. Data from the 2000-2001 school were used because these students we re in the tenth grade during that time period, which represents the last year of compulsory school attendance for many students. The disciplin ary referrals occurred at 47 different school sites, including traditional high schools, juvenile justice programs, exceptional student education centers, hospi tal/homebound education, vocational centers, and alternative education sites. Four referrals also occurred at middle school sites. Table 1 summarizes the racial demogr aphics of the sample used in the present study. The data indicate that the sample in this study was primarily Caucasian. Table 1 Numbers of African American and Caucas ian Students in the 1999-2000 Tenth Grade Sample with One Referral Category Frequency Percentage African American 449 26.9% Caucasian 1,218 73.1% Total 1,667 100%

PAGE 28

22 Table 2 contains information about t he free/reduced lunch status of the students in the tenth grade sample with one referral. Free/reduced lunch status served as an index of socioeconomic status in the present investigation. The majority of students in this sample we re not participatin g in the free/reduced lunch program. Table 2 Free/Reduced Lunch Status of Students in the 1999-2000 Tenth Grade Sample with One Referral Category Frequency Percentage Free or Reduced Lunch 360 21.6% Did not Apply 1,307 78.4% Table 3 summarizes the frequency of tenth grade students per number of referrals for the 1999-2000 school year. The distribution is skewed, with the majority of students at the lower end. The modal student received only one disciplinary referral duri ng the academic year. Ther efore, the present study focused on disciplinary actions applied to students with only one referral, thereby limiting the impact of multiple discipli nary referrals on consequences applied.

PAGE 29

23 Table 3 Frequency of Students Per Number of Referrals Number of Referrals Number of Students Percent 1 1667 42.9 2 442 11.4 3 328 8.4 4 239 6.1 5 159 4.1 6 158 4.1 7 130 3.3 8 97 2.5 9 78 2.0 10 69 1.8 11 72 1.9 12 54 1.4 13 50 1.3 14 33 0.8 15 27 0.7 16 39 1.0 17 35 0.9 18 24 0.6 Continued on the next page.

PAGE 30

24 Table 3 (Continued) 19 21 0.5 20 31 0.8 21 18 0.5 22 14 0.4 23 12 0.3 24 9 0.2 25 6 0.2 26 8 0.2 27 8 0.2 28 9 0.2 29 5 0.1 30 8 0.2 31 5 0.1 32 6 0.2 33 3 0.1 34 1 0.0 35 2 0.1 36 4 0.1 37 3 0.1 38 1 0.0 39 3 0.1 Continued on the next page.

PAGE 31

25 Table 3 (Continued) 40 1 0.0 41 1 0.0 42 1 0.0 43 0 0.0 44 0 0.0 45 3 0.1 46 0 0.0 47 0 0.0 48 2 0.1 49 2 0.1 50 0 0.0 51 0 0.0 52 0 0.0 53 0 0.0 54 0 0.0 55 0 0.0 56 2 0.1 Measures For this study, the independent variables were race (two levels: African American and Caucasian) and socioeconomic status (two sublevels: middle/high

PAGE 32

26 SES and low SES based on free/reduced lunch classification). The study’s dependent variable was type of administrative action taken (suspension or other). Because of the preponderance of literatur e indicating that African American students have been particularly affected by racial disproportionality in exclusionary discipline, only Caucas ian and African American students were included in the present analysis. Thus whether or not a student is suspended may partially depend on the student’s race. As stated previously, only data perta ining to students with one disciplinary referral for the 1999-2000 school year were included in the present analysis. This delimitation was selected because of the tendency of students with multiple administrative disciplinary contacts to be viewed and treated as “troublemakers” or subjected to harsher penalties for repeat offenses (Bowditch, 1993). To limit the potentially confounding effect of admin istrative policies and attitudes toward “repeat offenders,” only the disciplinary out comes of first-time referrals were analyzed. Overall, this study examined pos sible racial and socioeconomic differences in whether a student was suspended for five general types of offenses: property offenses, status o ffenses, controlled substances, violent offenses, and an “other” category us ed on the district referral forms. The school district in the present st udy classifies disciplinary referrals by referral codes, based on the type of offe nse. The district code of student

PAGE 33

27 conduct can be found in Appendix A. For th e purposes of this study, each of the offenses was classified into one of the four categories listed above, plus a fifth category called “other.” The “other” category encompasses referrals for which the exact offense code was not recorded. This classification system is adapted from Bowditch (1993) and provides a framework for simplifying and conceptualizing a lengthy list of offenses. Property offenses include such property-based acts as stealing and vandal ism. Status off enses include truancy and disrespect. These types of offenses are nonviolent and involve violations of social mores or rules specific to the sc hool setting rather than illegal acts. The controlled substances category encompa sses possession of alcohol, tobacco, and other unauthorized substances on school property. Violent offenses include fighting, possession of weapons, and a ssault. A separate analysis was conducted for each type of violation. The referral codes and their assigned categories are summarized in Table 4.

PAGE 34

28 Table 4 Referral Codes and Categories Referral Code Assigned Category Alcohol Controlled Substance Battery – Student Violent Battery – Adult Violent Drugs Controlled Substance Fighting Violent Leaving School Without Permission Status Disrespect – Defiance – Threats Status Profane/Obscene Language Status Repeated Misconduct Status Skipping Class Status Use of Tobacco Controlled Substance Stealing Property Weapons Violent Bus Misconduct Status Cheating Status Class Disruption Status Vandalism Property Missed Detention Status Other Other Continued on the next page.

PAGE 35

29 Table 4 (Continued) Missed Saturday School Status Excessive Tardies Status Lack of Cooperation Status In Unauthorized Area Status P.E. Misconduct Status Forgery Property Arson Property Breaking and Entering Property Sexual Harassment Status Threat/Intimidation Status Trespassing Property Electronic Devices Status Procedures Obtaining the data. The University of South Fl orida (USF) Institutional Review Board (IRB) evaluated the appropria teness of this study and determined that the procedures used in this study fu lfilled ethical standards. In addition to USF IRB approval, permission to access the data was obtained from the school district and project manager for the longitudinal study. Data Integrity. The data in these analyses were derived from school records of students enrolled in a Florida sc hool district. Preliminary examination

PAGE 36

30 was conducted to ensure the in tegrity of these archival data. Specifically, no student was included more t han once in the analysis. Type of Analysis. Descriptive statistics (fr equencies and percentages) were generated for each group examined in this study, where appropriate. These statistics were not only created for each group but also broken down by the five types of violations addressed in this study. Chi-square tests were conducted to ex amine the hypotheses presented in this study. Chi-square is a nonparametric statistical test to determine whether research data in the form of frequency counts are distributed differently in different samples. The first research question was exam ined using a chi-sq uare. Students receiving free/reduced lunch who had committ ed a violation tracked in the district database were compared with those who di d not receive free/reduced lunch in terms of total suspensions for each of the five types of offenses. The research has supported that SES is a major c onsideration when examining school outcomes such as suspension. This analysis evaluated whether or not significant differences existed betw een students from higher and lower SES groups in terms of their total suspensions. A second chi square analysis compared suspension rates of African American and Caucasian students for each of the five offense categories. The third question took SES into account, and compared Low SES students from both racial groups on the suspension outcomes of five types of offenses. A fourth set of chi-square analyses evaluated differences on the same offense outco mes in the middle to high SES group.

PAGE 37

31 Assumptions of Chi-Square. For a chi-square analysis, the sample must be randomly drawn from the population. In this study, the sample of tenth grade students with one disciplinary referral for the school year was analyzed. In addition, data must be reported in raw fr equencies, rather than percentages. The data set reported student demographics, refe rral codes, and disciplinary actions in terms of raw frequencies. Another assu mption of chi-square is that measured variables must be independent. No student was included more than once in the analysis conducted for this study. Also, no two values were generated from a single source. Values/categories on independent and dependent variables must be mutually exclusive and exhaustive. No student was counted as both Caucasian and African American, nor as both suspended and not suspended. Finally, observed frequencies cannot be too small. In this study, the expected values of each cell were consi dered likely to be greater than 5.

PAGE 38

32 Chapter IV Results This chapter describes the results of the study. The research questions presented in the Introduction guide the progr ession of this chapter. The total frequencies and percentages of referrals fo r each separate offense for all tenth graders are presented in Table 5, while fr equency data for the first-time referrals included in this analysis are presented in Table 6. T hese data are presented as more specific information about why students receive disciplinary referrals, beyond the five general categories exami ned in the Chi Square analysis. The most frequent offense was excessive tardie s, accounting for 20. 2% of referrals. When the related categories of ski pping class and leaving school without permission are added to this figure, att endance-related offenses accounted for 31.6% of the disciplinary referrals. T he categorically related offenses of disrespect, profane/obscene language, and lack of cooperation account for 24.9% of the total referrals. Status offenses accounted for the majority of referrals.

PAGE 39

33 Table 5 Referral Code Frequencies for All Tenth Graders Referral Code Frequency Percent Alcohol 15 0.1 Battery – Student 102 0.5 Battery – Adult 27 0.1 Drugs 107 0.6 Fighting 277 1.4 Leaving School Without Permission 592 3.1 Disrespect – Defiance – Threats 2288 11.8 Profane/Obscene Language 764 4.0 Repeated Misconduct 1029 5.3 Skipping Class 1607 8.3 Use of Tobacco 212 1.1 Stealing 63 0.3 Weapons 37 0.2 Bus Misconduct 208 1.1 Cheating 109 0.6 Class Disruption 1726 8.9 Vandalism 20 0.1 Missed Detention 1481 7.7 Other 1353 7.0 Continued on the next page.

PAGE 40

34 Table 5 (Continued) Missed Saturday School 905 4.7 Excessive Tardies 3908 20.2 Lack of Cooperation 1756 9.1 In Unauthorized Area 527 2.7 P.E. Misconduct 54 0.3 Forgery 51 0.3 Arson 5 0.0 Breaking/Entering 1 0.0 Sexual Offenses 1 0.0 Threat/Intimidation 78 0.4 Trespassing 3 0.0 Electronic Devices 9 0.0

PAGE 41

35 Table 6 Referral Code Categories for Students with One Referral Category Frequency Percentage Violent 155 9.3 Status 1223 73.4 Property 26 1.6 Controlled Substance 29 1.7 Other 218 13.1 Missing Data 16 1.0 Total 1667 100 Research Question 1. Are students identified as having low SES suspended from school more frequently than students not i dentified as having low SES? This question addressed w hether students receiving free and reduced lunch were suspended from school more frequently than stude nts not receiving free and reduced lunch for the same ty pes of offenses. Frequency data are reported in Table 7. The results of the chi square analyses are summarized in Table 8. Students receiving free and reduced lunch were significantly more likely to be suspended for status offenses, 2 (1, N = 1,223) = 8.74, p = 0.003.

PAGE 42

36 Table 7 Frequencies of Suspensions for Five Categories of Offenses, by SES Referral SES Suspended Not Suspended Violent Low 34 8 Violent High 80 33 Status Low 62 193 Status High 158 810 Property Low 4 5 Property High 10 7 Substance Low 2 1 Substance High 11 15 Other Low 9 41 Other High 19 149

PAGE 43

37 Table 8 Chi-Square Analysis for Research Ques tion 1, SES Comparison of Suspensions Referral Category Pearson’s Chi-Square Significance (2-sided) Violent 1.623 0.203 Status 8.737 0.003 Property 0.490 0.683 Substance 0.645 0.422 Other 1.541 0.215 Research Question 2. Are African American st udents suspended from school more frequently than Caucasian stud ents with disciplinary referrals for the same types of offenses? The results of the chi-square analyses are summarized in Table 10. African American students were significantly more likely than Caucasian students to be suspended from school for violent offenses, 2 (1, N = 155) = 9.70, p = 0.002 and status offenses, 2 (1, N = 1,223) = 36.47, p = 0.000, as well as offenses classified as “other,” 2 (1, N = 218) = 9.13, p = 0.003.

PAGE 44

38 Table 9 Frequencies of Suspensions for Five Categories of Offenses, by Race Referral Race Suspended Not Suspended Violent AA 57 9 Violent C 57 32 Status AA 93 226 Status C 127 777 Property AA 7 6 Property C 7 6 Substance AA 1 0 Substance C 12 16 Other AA 12 34 Other C 16 156 AA = African American, C = Caucasian

PAGE 45

39 Table 10 Chi-Square Analysis for Research Ques tion 2, African American and Caucasian Students Referral Category Pearson’ s Chi-Square Significance (2-sided) Violent 9.703 0.002 Status 36.467 0.000 Property 0.000 1.000 Substance 1.275 0.259 Other 9.134 0.003 Research Question 3. Are low SES African American students suspended from school more frequently than low SES Caucasian students for the same types of offenses? The results of the chi-square analyses are summarized in Table 12. The results of the chi-squar e analyses do not indicate significant significant racial differences in sus pensions for low SES students referred for any of the five types of offenses.

PAGE 46

40 Table 11 Frequencies of Suspensions of Low SES Students for Five Categories of Offenses, by Race Referral Race Suspended Not Suspended Violent AA 23 3 Violent C 11 5 Status AA 42 109 Status C 20 84 Property AA 2 4 Property C 2 1 Substance AA 0 0 Substance C 10 0 Other AA 7 19 Other C 19 149

PAGE 47

41 Table 12 Chi-Square Analysis for Research Ques tion 3, Low SES African American and Caucasian Students Referral Category Pearson’s Chi-Square Significance (2-sided) Violent 2.496 0.114 Status 2.466 0.116 Property 0.900 0.343 Substance not computed* N/A Other 2.922 0.87 *Insufficient number of cases Research Question 4. Are middle and higher SES African American students suspended from school more fr equently than middle and higher SES Caucasian students for the same types of offenses? The results of the chisquare analyses are summarized in Table 14. Middle and higher SES African American students were significantly more likely to be suspended from school for status offenses 2 (1, N = 968) = 29.32, p = 0.000 and violent offenses 2 (1, N = 113) = 6.04, p = 0.014.

PAGE 48

42 Table 13 Frequencies of Suspensions of Middle and Higher SES Students for 5 Categories of Offenses, by Race Referral Race Suspended Not Suspended Violent AA 34 6 Violent C 46 27 Status AA 51 117 Status C 107 693 Property AA 5 2 Property C 5 5 Substance AA 1 0 Substance C 10 15 Other AA 5 15 Other C 14 134

PAGE 49

43 Table 14 Chi-Square Analysis for Middle and Higher SES Students (African American and Caucasian) Referral Category Pearson’s Chi-Square Significance (2-sided) Violent 6.042 0.014 Status 29.317 0.000 Property 0.781 0.377 Substance 1.418 0.234 Other 4.212 0.39 Analysis at the School Level. Due to the large number of referrals countywide and the potential for variability among school sites, the referrals were analyzed at the individual school leve l. Table 15 presents the frequency and percentage of referrals from the four schools with the highest numbers of single referrals in the study. These high school s are identified as High Schools A, B, C and D. These sites accounted for a total of 466 referrals, or a combined 27.9% of the 1667 referrals included in this analysis.

PAGE 50

44 Table 15 Frequency and Percentage of Single Referrals, by Site Site Frequency Percent High School A 135 8.1 High School B 114 6.8 High School C 107 6.4 High School D 110 6.6 Tables 16 through 19 provide information about the frequency and percentage of African American and Caucasian student s with single referrals suspended for each of the five offense categories, at each of the four target high schools.

PAGE 51

45 Table 16 Frequency and Percentage of Suspensions for 5 Referral Categories by Race, High School A Referral Race Frequency Percent Violent AA 5 100% Violent Caucasian 5 83.3% Status AA 5 11.4% Status Caucasian 11 19.0% Property AA 1 100% Property Caucasian 1 100% Substance AA 0 0% Substance Caucasian 3 75% Other AA 2 40% Other Caucasian 1 9.1%

PAGE 52

46 Table 17 Frequency and Percentage of Suspensions for 5 Referral Categories by Race, High School B Referral Race Frequency Percent Violent AA 6 85.7% Violent Caucasian 7 77.8% Status AA 7 31.8% Status Caucasian 8 11.8% Property AA 1 100% Property Caucasian 0 0% Substance AA 0 0% Substance Caucasian 0 0% Other AA 0 0% Other Caucasian 5 100%

PAGE 53

47 Table 18 Frequency and Percentage of Suspensions for 5 Referral Categories by Race, High School C Referral Race Frequency Percent Violent AA 4 80% Violent Caucasian 6 85.7% Status AA 7 23.3% Status Caucasian 7 13.7% Property AA 1 100% Property Caucasian 1 100% Substance AA 0 0% Substance Caucasian 1 100% Other AA 0 0% Other Caucasian 1 14.3%

PAGE 54

48 Table 19 Frequency and Percentage of Suspensions for 5 Referral Categories by Race, High School D Referral Race Frequency Percent Violent AA 1 100% Violent Caucasian 5 83.3% Status AA 4 23.5% Status Caucasian 14 20.0% Property AA 0 0% Property Caucasian 1 100% Substance AA 0 0% Substance Caucasian 1 100% Other AA 0 0% Other Caucasian 1 7.1% Table 20 summarizes the percentages of African American and Caucasian students with each referral code who were suspended from school.

PAGE 55

49 Table 20 Percentages of Caucasian and African Am erican Students Suspended for Each Referral Code Referral Code AA Caucasian Alcohol N/A* 100% Battery – Student 78.9% 50% Battery – Adult 85.7% 100% Drugs 75% 100% Fighting 87.9% 64.9% Leaving Campus 72.7% 25.9% Disrespect/Defiance/Threats 43.9% 25.7% Profane/Obscene Language 45.0% 35.6% Repeated Misconduct 62.5% 33.3% Skipping Class 0% 2.8% Use of Tobacco 100% 23.8% Stealing 55.6% 71.4% Weapons 100% 100% Bus Misconduct 18.2% 27.3% Cheating 0% 9.1% Class Disruption 27.5% 13.8% Continued on the next page.

PAGE 56

50 Table 20 (Continued) Vandalism N/A* 100% Missed Detention 4.8% 5.0% Other 26.1% 9.3% Missed Saturday School 100% 80% Excessive Tardies 100% 100% Lack of Cooperation 16.7% 5.1% In Unauthorized Area 10.0% 4.7% P.E. Misconduct N/A 100% Forgery 33.3% 33.3% Sexual Harassment N/A 50.0% Threat/Intimidation 100% 68.7% Electronic Devices N/A 100%

PAGE 57

51 Chapter V Discussion The purpose of this study was to inve stigate racial equity in the use of suspension and expulsion among tenth gr ade students in a large Florida school district. The study was conducted using archival discipline data from the general student data files. Information used in the analysis included the student’s race, referral code, disciplinary action code, and free and reduced lunch status. This chapter will address each research hypot hesis, including how each hypothesis was supported, implications of the findings and directions for future research. Results of this study will be synthesized and in terpreted. Limitations of the study will be presented, as well as potential ques tions to be addressed through future research. Implications for school psyc hology practitioners also will be discussed. Hypothesis 1. Students identified as havin g low socioeconomic status (SES) for free/reduced lunch stat us will be suspended from school more frequently than students not ident ified as having low SES. a. Low SES students will be suspended fr om school more frequently than middle and higher SES students for violent offenses. b. Low SES students will be suspended fr om school more frequently than middle and higher SES students for property offenses. c. Low SES students will be suspended fr om school more frequently than middle and higher SES student s for status offenses.

PAGE 58

52 d. Low SES students will be suspended fr om school more frequently than middle and higher SES students for offenses involving controlled substances. e. Low SES students will be suspended fr om school more frequently than middle and higher SES students for of fenses categorized as “other.” The first set of hypotheses were test ed through chi square analysis. This hypothesis was supported for one category of disciplinary referral. Low SES students were more likely to be suspended fr om school for status offenses, such as skipping class, than their middle and higher SES peers. They were not significantly more likely to be suspended for property offenses such as stealing or vandalism, violent offenses, offenses in volving controlled substances, or those falling into the “other” category. The impact of school performance and grades should be considered when interpreting these results. Low SES students often receive lower grades and experience more academic difficult ies than middle and higher SES students (Bowditch, 1993). Students who receiv e poor grades and experience many academic difficulties may derive less posit ive reinforcement for attending school, participating in classroom activities, and following the rules. Such difficulties could be related to a higher likelih ood of skipping class and subsequent suspension. Hypothesis 2. African American students will be suspended from school more frequently than Caucasian students.

PAGE 59

53 a. African American students will be suspended from school more frequently than Caucasian students for violent offenses. b. African American students will be suspended from school more frequently than Caucasian students for property offenses. c. African American students will be suspended from school more frequently than Caucasian studen ts for status offenses. d. African American students will be suspended from school more frequently than Caucasian students for offenses involving controlled substances. e. African American students will be suspended from school more frequently than Caucasian students for of fenses categorized as “other.” This hypothesis was supported for two referral categories, status and violent offenses. The resu lts of the chi-square analysis indicate that African American students who committed status offenses were more likely to be suspended than Caucasian students w ho committed status offenses. Additionally, African American students who we re referred to the office for violent offenses were more likely to be suspended than Caucasian students with the same referral category. Hypothesis 3. Low SES African American students will be suspended from school more frequently than Low SES Caucasian students. a. Low SES African American studen ts will be suspended from school more frequently than Low SES C aucasian students for violent offenses.

PAGE 60

54 b. Low SES African American studen ts will be suspended from school more frequently than Low SES C aucasian students for property offenses. c. Low SES African American studen ts will be suspended from school more frequently than Low SES Caucas ian students for status offenses. d. Low SES African American studen ts will be suspended from school more frequently than Low SES C aucasian students for offenses involving controlled substances. e. Low SES African American studen ts will be suspended from school more frequently than Low SES C aucasian students for offenses categorized as “other.” This hypothesis was not supported for any referral category. The chi square analysis indicated that low SES African American students were not significantly more likely than low SES Caucasian students to be suspended from school for any of the five types of offenses included in the analysis. Hypothesis 4. Middle and Higher SES Afri can American students will be suspended from school more frequently than Middle and Higher SES Caucasian students. a. Middle and higher SES African American students will be suspended from school more frequently than middle and higher SES Caucasian students for violent offenses.

PAGE 61

55 b. Middle and higher SES African American students will be suspended from school more frequently than middle and higher Caucasian students for property offenses. c. Middle and higher SES African American students will be suspended from school more frequently than middle and higher SES Caucasian students for status offenses. d. Middle and higher SES African American students will be suspended from school more frequently than middle and higher SES Caucasian students for offenses involv ing controlled substances. e. Middle and higher SES African American students will be suspended from school more frequently than mi ddle and higher SES Caucasian students for offenses cat egorized as “other.” Similar to the pattern seen when SES wa s not considered, this hypothesis was supported for status and violent o ffenses only. Middle and higher SES African American students were significantly more likely to be suspended for status offenses and violent offenses than their Caucasian peers with referrals of the same variety. It appears that whether or not a student is suspended from school depends on a combinatio n of factors: the nature of the offense, race, and socioeconomic status. Comparisons within schools. Due to the small numbers of students with single referrals in each category, chi-square analys is at the individual school level was not feasible. However, examination of raw frequency and percentage data for four high-referring high schools provi des insight into how discipline is

PAGE 62

56 implemented at these sites. For each of the four high schools, African American students with one referral were more likely to be suspended for status offenses than their Caucasian peers who committed t he same category of offense. In three out of four sites, African Amer ican students who committed violent offenses also were more likely to be suspended than Caucasian students with the same referral category. It is difficult to produce broad gener alizations from the data about individual schools, because the sample size is small and further restricted by the delimitations of this study. For ex ample, students with multiple disciplinary referrals were excluded from the pres ent study. However, when addressing discipline issues at the school level, it would be important to include data regarding the students most likely to co me in contact with administrative discipline. Summary and Implications Previous research has extensively documented the overr epresentation of African American students among those su spended and expelled from school. The present study sought to investi gate whether, given the same general category of offense, African American st udents were more likely to be suspended or expelled from school than Caucasian stud ents. The results indicate that, for the group of students included in this analysis, racial equity was related to the type of offense, as well as to the st udent’s socioeconomic status. When race was not considered, low SES students referred for status offenses were more likely to be suspended than middle and higher SES students with the same

PAGE 63

57 referral type. However, Low SES African American students were not significantly more likely to be suspended from school than low SES Caucasian students with the same type of referral, for any of the fi ve offense categories. Racial differences were found when SES was not considered, with African American students more likely to be suspen ded from school for status offenses and violent offenses. The same degree of racial disproportionality was not found among low SES students. However, middle and higher SES students appeared to account for much of the racial di sproportionality seen in the sample, with African American students in this group more likely to be suspended for both violent and status offenses. The results of this study indicate that African American students remain overrepresented among those suspended and expelled for offenses involving social infractions (e.g., verbal disrespec t, truancy) and those involving violence. These findings support those of Skiba et al. (2000), in which race was strongly related to the use of suspension. However, it is important to no te that low SES is a significant risk fact or for suspension, regardless of race. The findings of this study echo previous research (Costenbader & Markson, 1994) in identifying violent off enses as those lending themselves most readily to suspension. These findings ma ke sense in light of zero tolerance policies and the overall need to preser ve safety in the school environment. The relationship between SES and suspension from school may be related to a number of factors. Student s of lower socioeconomic status often obtain lower grades and achievement test scores than their middle and higher

PAGE 64

58 SES counterparts (Mujis, 1997). In addition, some of these students may have adopted values inconsistent with the pursuit of higher education and mainstream conformity (Bowditch, 1993). Low SES students also may lack the strong parental advocacy that is often helpful in negotiating with school personnel, thereby remaining more vulnerable to ex clusionary discipline. Finally, low SES students may be impacted by referral bias in the classroom. Limitations of this Study The present study is limited by the fact that it is a secondary analysis. As such, the author did not have control over how the data were collected and maintained. Inclusion of the “other” category is es pecially problematic, because no information is provided about the actual referral offense for these cases. In addition, free and reduced lunch status is a crude measure of socioeconomic status. Some low-income families do not apply for free and reduced lunch, and there is certainly income variation within ea ch of the two lunch status groups. In addition, lunch status provides no di rect information about parental education, family orientation toward school, or parenting strategies. These types of information would provide additional insight into the factors that shape students’ behavior and attitudes toward school and ma y subsequently impact disciplinary outcomes. The low numbers of referrals fo r property and controlled substance offenses present a problem in the pres ent analysis, as they are too low to be meaningfully interpreted. To fully under stand how exclusionary discipline is used for these types of offenses, a lar ger sample size would be necessary.

PAGE 65

59 Another limitation is that these data do not allow examination of the finer nuances surrounding various offenses, such as the specific type of assault or the nature of disrespectful conduct, the num ber of students involved, where the incident occurred, or behavioral interv entions (if any) attempted prior to administrative referral. Certain delimitations were placed upon the study, in order to focus specifically on the problem of interest. Only Caucasian and African American students were compared, because the impac t of exclusionary discipline on the educational experience of the Af rican American population has been documented in the extant literature as a serious concern. However, because of this delimitation the result s of the present study c annot be generalized to other ethnic groups. Future research shoul d address the use of suspension with Latino, Native American, and language mi nority students. St udents with multiple disciplinary referrals were excluded from this analysis, because of the complex impact of administrative attitudes and policies, students’ r eputations, and other factors on the disciplinary outcomes of “serial offenders.” These results cannot be generalized to students with a long history of disciplinary infractions leading to numerous administrative referrals. Directions for Future Research The present study demonstrated t hat African American students were overrepresented among those suspended fr om school for violent and status offenses, regardless of socioeconomic stat us. While this study elected to focus only on Caucasian and African American st udents, the diversity of many public

PAGE 66

60 school systems warrants investigation of the use of exclusionary discipline with other racial, cultural, and language mi nority groups. Thus, future research should examine the use of suspension wi th other racial and ethnic groups. Referral bias is another important issue to be explored in future studies. Given the results obtained and theories posit ed by such researchers as Skiba et al. (1999) and Costenbader and Markson (1994), it could be predicted that African American students will be referr ed to the office more often than Caucasian students committing the same type s of infractions. This hypothesis could be examined through naturalistic observation of teacher responses to classroom behavior with students of differ ent races at the middle and high school levels. In light of the present results, socioeconomic status also should be taken into account. While this type of study w ould likely necessitate a smaller sample size than the present investi gation, it also would allow for direct observation of why students may or may not be referred to the office for a given type of offense. In addition, the impact of multiple su spensions from school is an important issue for future inquiry. Extant res earch has demonstrated that previous suspensions from school are a strong predi ctor of future suspensions (Raffaele Mendez, 2003). Therefore, racial equity in the use of multiple suspensions also must be addressed, as well as the impac t of multiple suspensions on a student’s educational trajectory. Future research could address this issue by examining how suspension is applied with different racial and ethnic groups beyond the first administrative referral. In addition, such important outcomes as high school graduation rates, higher educational attain ment, juvenile justice involvement,

PAGE 67

61 substance abuse, and teen par enthood should be examined relative to the use of exclusionary discipline. Also, the fa ct that nearly half of all students with disciplinary referrals received only one for the entire academic year raises the allimportant question of why some students continue to receive referrals while others do not. Future research should de lve into the risk and protective factors that differentiate students with one referral from repeat offenders. Finally, the efficacy of sc hool-based behavior management and intervention programs shoul d be compared with suspension for various ethnic groups. Alternatives to suspension warr ant extensive research, because current research does not support the efficacy of suspension as a strategy for behavior change. Exclusionary discipline is generally an ineffective intervention, as it is associated with the need for repeated sus pensions (Raffaele Mendez, 2003). There is a need for practical, empiricall y supported interventions that provide education and remediation, rather than exclusion, for students with problem behaviors. Implications for School Psychology Practitioners Researchers and practitioners have l ong questioned the effectiveness of suspension and expulsion as helpful in terventions for students with serious problem behaviors. However, exclusionar y discipline remains prevalent in many school systems. From an administrativ e viewpoint, suspension and expulsion are often seen as means for mainta ining a safe and orderly educational environment (Wu et al., 1983). As child advocates, school psychologists are often faced with a difficult “balancing ac t” when administrators insist that a

PAGE 68

62 student with problem behaviors be removed from the school setting. Real and exaggerated fears of school violence cont ribute to the overuse of suspension, at the expense of students’ access to a fr ee and appropriate public education. However, school safety and maintenanc e of an environment conducive to learning for all students remain i ssues of critical importance. Chronic and serious problem behav iors must be addressed in a manner that promotes positive outcomes for all students. When a child exhibits chronic problem behaviors that are resistant to intervention strategies used in the classroom, or when a child presents a seri ous threat of harm to self or others, remedial programs present a preferable alternative to suspension. Effective remedial discipline programs share a num ber if characteristics, summarized by Bear, Cavalier, & Manning (2002). First, they target multiple risk and protective factors within a comprehensive framework. These programs also adopt a broadbased, ecological systems perspective that facilitates collaboration among schools, agencies, and families. Effectiv e remedial programs for aggressive and antisocial behavior also use empirically s upported interventions in a manner that remains intensive and sustained over a signi ficant period of ti me. Additionally, effective programs are sensitive to the developmental appropriateness of intervention strategies. Finally, thes e programs include early intervention strategies for targeting problem behavio rs at an early age or at the beginning stage of problem development. School psychologists are in a position to potentially influence school and public policy regarding the treatment of children with chronic and severe problem

PAGE 69

63 behaviors. School staff training and di ssemination of information regarding alternatives to exclusionary discipline ar e important roles. In addition, ongoing program evaluation at the school and district levels is important for obtaining information about which interventions are effective and which need to be modified. Funding and support at the government level for remedial programs also will increase the likelihood that in tensive intervention strategies can be maintained over time with some degree of integrity. The results of this study indicate that racial inequity remains a serious problem with the use of exclusionary discipline. Suspension from school represents the denial of educational opportunities and resources to the students who are most in need.

PAGE 70

64 References Advancement Project/Civil Rights Educ ation Project (2000, February). Education denied: The negative impact of zero tolerance policies. Testimony before the United States Commission on Civ il Rights, Washington, D.C. Bowditch, C. (1993). Ge tting rid of troublemakers: High school disciplinary procedures and the production of dropouts. Social Problems, 40 (4), 493508. Cooper, J. O., Heron, T. E. & Heward, W. L. (1987). Applied Behavior Analysis. Merril: Upper Saddle River, NJ. Costenbader, V., & Markson, S. (1994). Sc hool suspension: A survey of current policies and practices. NASSP Bulletin, 103-113. Costenbader, V., & Markson, S. (1998) School suspension: A study with secondary school students. Journal of School Psychology, 36 (1), 59-82. Gordon, R., Piana, L. D., & Keleher, T. (2000). Facing the consequences: An examination of racial discriminat ion in U.S. public schools. Gregory, J. F. (1997). Th ree strikes and they’re out: African American boys and American schools’ responses to misbehavior. International Journal of Adolescence and Youth, 7, 25-34. Johnston, R. C. Federal data high light disparities in discipline. Education Week. Available online: http://edweek.org. Keleher, T. (2000). Racial disparities related to school zero tolerance policies. Testimony before the U.S. Commission on Civil Rights, Washington, D.C.

PAGE 71

65 MacMillan, D. L., & Reschly, D. J. ( 1998). Overrepresentation of minority students: The case for greater specif icity or reconsideration of the variables examined. Journal of Special Education, 32, 154-162. McFadden, A. C., Marsh, G. E. Price, B. J., & Hwang, Y. (1992). A study of race and gender bias in the punishment of school children. Education and Treatment of Children, 15, 141-146. Morgan-D’Atrio, C., Northup, J., LaFleur, L., & Spera, S. (1996). Toward prescriptive alternatives to sus pensions: A preliminary evaluation. Behavioral Disorders, 21 (2), 190-200. Morrison, G. M. & Skiba, R. J. (2001). Predicti ng violence from school misbehavior: Promises and perils. Psychology in the Schools, 38 (2), 173184. Mujis, R. D. (1997). Predictors of academic achievement and academic selfconcept: A longitudinal perspective. British Journal of Psychiatry, 67 (3), 263-277. New York State Attorney General’s Office-Civil Rights Bureau (1999). The New York City Police Department’s stop & fri sk practices: A report to the people from the State of New York Of fice of the Attorney General. New York: Author. Office for Civil Rights (1994). 1992 Elementary and Secondary School Compliance Report: Final Fi le User’s Guide. U.S. Department of Education; Washington, D.C.

PAGE 72

66 Osler, A., Watling, R., Bus her, H., Cole, T., & White, A. Reasons for exclusion from school. Research Brief No 244, Department for Education and Employment, London. Raffaele Mendez, L. M. (2003, May). Predictors of suspension and suspension as a predictor of negative school outcome s: A longitudinal investigation of an entire cohort of student s from second through twelfth grades. Paper presented at the School to Prison Pi peline Conference sponsored by the Civil Rights Project, Harvard University, Boston, MA. Raffaele Mendez, L. M., & Knof f, H. M. (2003). Who gets suspended from school and why: A demographic analysis of school s and disciplinary infractions in a large district. Education and Treatment of Children, 26 (1), 30-51. Reschly, D. J. (1997). Dis proportionate minority representation in general and special education: Patterns, issues, and alternatives. Des Moines, IA: Mountain Planes Regional Resource Center, Drake University. Rodney, L. W., Crafter, B., Rodney, H. E. & Mupier, R. M. (1999). Variables contributing to grade ret ention among African Americ an adolescent males. Journal of Educational Research, 92 (3), 185-190. Rose, T. (1988). Current discipli nary practices with handicapped students: Suspensions and expulsions. Exceptional Children, 55, 230-239. Serwatka, T. S., Deering, S., & Grant, P. (1995). Disproportionate representation of African Americans in emot ionally handicapped classes. Journal of Black Studies, 25, 492-506.

PAGE 73

67 Skiba, R. J., Michael, R. S ., Nardo, A. C, & Patterson, R. (2000). Sources of racial and gender disproportionality in di scipline. Policy Research Report #SRS1, Indiana Educat ion Policy Center. Skiba, R. J., Peterson, R. L., & W illiams, T. (1997). Office referrals and suspensions: Disciplinary referrals in middle schools. Education and Treatment of Children, 20 (3), 295-315. Sprague, J., Walker, H. M., Stieber, S., Si monsen, B., Nishioka, V., & Wagner, L. (2001). Exploring the relationship bet ween school discipline referrals and delinquency. Psychology in the Schools, 38 197-206. Uchitelle, S., Bartz, D., & Hillman, L. (1989). Strategies for reducing suspensions. Urban Education, 24 (2), 163-176. Wu, S. C., Pink, W., Crain, R., & Moles, O. (1982). Stu dent suspension: A critical reappraisal. The Urban Review, 14 (4), 245-273.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001498132
003 fts
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 041209s2004 flu sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000535
035
(OCoLC)57717271
9
AJU6727
b SE
SFE0000535
040
FHM
c FHM
090
LB1051 (ONLINE)
1 100
Tremper, Mary M.
0 245
Racial equity in exclusionary discipline practices
h [electronic resource] /
by Mary M. Tremper.
260
[Tampa, Fla.] :
University of South Florida,
2004.
502
Thesis (Ed.S.)--University of South Florida, 2004.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 73 pages.
520
ABSTRACT: The present study examined whether external (out-of-school) suspensions are applied equitably to students of different ethnic backgrounds who commit violent and nonviolent offenses. The hypotheses presented in this study were addressed through secondary analysis of disciplinary records from a large metropolitan school district in Florida. The results indicate that, for the group of 1,667 tenth grade students included in this analysis, racial equity was related to the type of offense, as well as to the student's socioeconomic status. Racial differences were found when SES was not considered, with African American students more likely to be suspended from school for status offenses and violent offenses. The same degree of racial disproportionality was not found among low SES students.However, middle and higher SES students appeared to account for much of the racial disproportionality seen in the sample, with African American students in this group more likely to be suspended for both violent and status offenses.
590
Adviser: Powell-Smith, Kelly.
653
race.
suspension.
socioeconomic status.
behavior management.
adolescents.
690
Dissertations, Academic
z USF
x School Psychology
Specialist.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.535